TRANSISTORS NEEDED

[AudioSan]

so it is the same problem with Firefox and Explorer.

 

[bnorrish]

I had a Firefox funeral recently. I have used for years but had a lot of issues with recent builds. nVidia kernal driver crashes on Windows 8 was the last straw. I run Chrome unless accessing one of my corporate applications at work where I need to use Internet Explorer. Chrome is lean and fast - works.

 

[6L6]

You should be able to measure Idss, gm and noise. .

I know how to test Idss.

How to test Noise and gm?

 

[buzzforb]

What if the new LSZkdo not perform as well as the NOS. Still one of only s few options other than going to bjt's or mosfets.

 

[6L6]

The LS are indistinguishable from the Toshiba, as far as I can tell.

And as my last 3 amps have Toshiba driving one channel and LS driving the other, I can't tell a difference looking at the output, and that's all that matters to me.

 

[bear]

The BDENT J74s are NG, imo.

They curve trace like the asian bogus parts.

I think they got taken too...

_-_-

 

[6L6]

Really? That sucks.

Mine are supposed to arrive today.

 

[bear]

I am pretty confident in saying that there are many F5 amps that have been built with the bogus J74s from asia by now! Probably a majority of them!! Likely very good for enhancing the even order harmonic distortion.

_-_-

 

[bear]

Really? That sucks.

Mine are supposed to arrive today.

Quite so. But miracles do occur.

They were a great source of imported semiconductors for years and years, so I bought a large amount of them about a year or so ago... NG. I was hoping... but you never know what batch they bought or you get...

Curve tracing is a must. They will meet Idss.

There's a $20 curve tracer on ebay that hooks to ur scope, ought to be good enough. (may have to convert to voltage drive though...)

_-_-

 

[kasey197]

I don't know of many (any?) p channel jfets that have the combination of idss and yfs that the SJ74s have. Noise is another matter again

If these are fakes, then whatever part it is that the fakers are using is tough to nail down ....

 

[buzzforb]

could they be clones o throw aways?

 

[bear]

the other thread on bad devices has examples with images, it is not that hard to find out which devices have been rebadged... (in this case)

and they do not match the K170s at all except for IDSS.



_-_-

 

[buzzforb]

I have bought exclusively from Spencer. I have traced all but the most recent batch, but have never seen anything unusual. The last batch looks visibly good in comparison, but have not been traced.

 

[labjr]

I really wish the LS parts would become available soon so we'll know if they're any good or not and can plan before there are no Toshiba parts anymore.

 

[jackinnj]

I know how to test Idss.

How to test Noise and gm?

Linear Systems has archived Erno Borbley's papers on testing JFETs and the math for the gm test is in there:

http://www.linearsystems.com/assets... The New Frontier Part 1, by Erno Bordely.pdf

This paper at GA Tech (the late Marshall Leach perhaps) shows a method of measuring noise:

http://users.ece.gatech.edu/mleach/ece6416/Labs/newexp5a.pdf

The noise from a K170/J74 is going to be less than most DIYr instruments noise floor -- under 1nV/Rt Hz -- so you set up the device as a amplifier with gain of 20 or 40dB, a.c. couple it to a low noise opamp like the LT1028 or AD797 and filter the output. (A device like the Quan-Tech 5173 uses a series of steep bandpass filters to arrive at nV/RtHz).

Dennis Colin also had a couple articles in AudioXpress on the topic.
http://audioamateur.com/files/2009/02/colin2993.pdf

You can drive yourself to distraction with these measurements.

 

[henryve]

I bought 200 2SK170BLs from Zhoufang. I hope his stuff is still legit. My 2SJ74BLs I got from Spencer.

 

[Triodethom]

claims to have real stock any feed back ? 2SJ74BL PC-TOSHIBA FET- AF DRIVER - BL-ENDING ORIGINAL

 

[henryve]

claims to have real stock any feed back ? 2SJ74BL PC-TOSHIBA FET- AF DRIVER - BL-ENDING ORIGINAL

Yes, but the table on this page makes me wonder.

Manufacturer = N/A ?
Description = PC-FET- AF DRIVER - Replacement .. ?

I wouldn't trust that.

 

[jammer55]

The 2sj74bl I bought from them were fake. Yfs on the j74s from PacificSemi were all around 8ms. The 2sj104gr I got from them all measured above 26ms.
I think I got the last few j104 of their stock. Measurements were taken with a Peak Atlas DCA75.

I also got fake 2sj107 from hkutsource at ebay. I informed them, with measeurements, that the parts were fake. They haven't removed the listing.

 

[Triodethom]

I ordered a handful of the matches, hopefully they are genuine. Will have them by the end of the week.

How is the testing going?